In the 1980s, a new type of industrial robot (parallel robot) with a parallel mechanism as its main structure continuously provided more perfect solutions for certain industrial fields, attracting widespread attention from industry and academia.
Among the various types of parallel robots, there exists a category of parallel robots (also known as parallel manipulators) driven by external rotary/prismatic joints and containing parallelogram branches. Because the drive unit can be mounted on a static platform (frame), and the driven arm is often made of lightweight, thin rods, the end effector platform can achieve very high speeds and accelerations. This makes them particularly suitable for sorting, handling, and gripping operations on high-speed logistics production lines, and thus they have gradually become a hot topic in research and development in recent years.
Research on high-speed parallel robots can be traced back to the Delta manipulator invented by Dr. Clavel in 1985. This robot's active arm is driven by an external revolute joint, while the driven arm has a parallelogram structure. The end effector can achieve high-speed 3D translation within the workspace. Furthermore, by adding a retractable pivot with Hooke hinges at both ends between the static and moving platforms, the end effector can achieve single-degree-of-freedom rotation about the plane normal to the moving platform, thus completing the grasping and releasing action on the target object.
Parallel robots have to overcome numerous obstacles to move from demand to actual market application; it's a problem that must be solved!
To create a parallel robot that meets market demands, the fundamental challenge lies in mastering the technical essentials of high-speed, high-precision operation using high-speed parallel robotic arms! This is also a core issue for parallel robots, in addition to the three core technologies of robotics (controllers, servo motors, and reducers). Let's delve into the technical challenges of parallel robots with Atom Robotics, a leading company in this field.
First Challenge—Topology Synthesis
Laymen see the spectacle, experts see the details. Many newcomers or those just starting out in this field are undoubtedly thinking, "Is the first hurdle really this difficult? What is this picture full of circles? I don't understand it!"
This technique, called topological synthesis, primarily aims to invent mechanisms that meet specific motion requirements. The parallel robots we see today initially all looked like this. The mathematical community has a comprehensive approach to solving this problem, including contractile graph theory, combinatorial analysis, enumeration theory, and spiral theory. After reading this, one can only marvel at the power of mathematics! Without this powerful mathematical approach, we wouldn't have been able to overcome this first hurdle! Keep up the good work, parallel robot engineers!
The second challenge—parameter design
This looks like the shape of a robot, but why are there so many symbols? Completely baffled. We've passed the first hurdle and obtained a mechanism configuration that roughly meets the requirements, but it's unusable! How do we know the mathematical relationship between the input and output? How do we determine the dimensional parameters of each component? How do we determine the robot's size? Furthermore, how do we train the mechanism to develop "muscles" and determine the structural parameters of each component? And how do we determine the appropriate force to drive the mechanism's movement?
Technically, the first step is to establish a kinematic and dynamic model describing the input-output relationship of our mechanism using spatial vectors or spiral theory. The second step involves extracting the optimization objectives and constraints from the kinematic and dynamic equations based on the mechanism's motion characteristics, and then optimizing the mechanism's dimensional and structural parameters. The third step is to determine the mechanism's driving parameters based on these dimensional and structural parameters using mathematical tools such as Newton and Euler. In fact, the process is as follows:
Okay! Even a newbie like me finds it mysterious enough!!! But at least I finally made it through the second level by carefully taking each step!
The third stage—virtual simulation
Astro Boy robots are targeting the global market! Therefore, meeting domestic and international standards and conducting experiments with realistic data are prerequisites for enterprise production! Thus, the third hurdle must be overcome - virtual simulation.
At this point, the company needs to master all the skills of MATLAB, Adams, Samcef, and ANSYS! Only after using this combination and overcoming numerous challenges can we proudly say that there are no problems and that the simulation analysis matches the theoretical design.
Now that we've passed the simulation analysis stage, can we start drawing the part blueprints? Of course not!
Do you know how to assign tolerances and precision for each part? Do you know which parts require special attention? Do you know how to ensure the assembly precision of components? Do you know the assembly process for components? No? Then you'd better figure that out quickly!
The fourth stage—precision design and calibration
Finally, after many hardships, we arrived at the fourth level, which is to establish an error analysis model for the robot and separate compensable errors from uncompensable errors.
For compensable errors, we need to establish a kinematic calibration model and perform compensation in the subsequent control system; for uncompensable errors, we need to conduct sensitivity analysis and tolerance design to determine the manufacturing and assembly accuracy of key components, and determine the assembly process and testing methods of the components based on this accuracy.
Seeing this, many people finally understand and will never again say that the core technology is lacking. But what if they don't understand these things? They have to learn! This presents yet another series of mathematical challenges!
Level 5 - Controllers and Algorithms
After passing these four hurdles, it felt like a rainbow had appeared in the sky; finally, we could look forward to building the prototype. However, the happy moments were fleeting. It seems we've missed something? Yes, we've only built the skeleton and muscles of the parallel robot; how do we build the brain that controls its behavior? How does it form its thoughts?
Fortunately, we could buy the brain from abroad, but how could we let the core technology be in someone else's hands? Besides, we still had excellent algorithms to incorporate! Therefore, the Astro Boy robot team made up their minds and, despite market opposition to independent research and development, established the mysterious Brain Research Department!
From then on, the department began a training mode that knew no day or night. Countless days and nights, countless boxes of ammunition, were spent in this place. There was no other way; our foundation was weak! We could only keep working hard and make up for our shortcomings through diligence! Fortunately, Astro Boy's only asset was that he was young, persevering, and good at learning.
After three years of continuous research and development, the Astro Boy robot has finally completed the construction of its brain! Breakthroughs have also been achieved at the cognitive level! Through research into motion principles and vibration suppression algorithms, the robot's precision has been further improved!
With victory seemingly within reach, many were undoubtedly filled with excitement, joy, and tears! They felt they could finally breathe a sigh of relief, but! Because we're making products, they need to undergo the most rigorous testing in the market! We can't just do research! This research needs to be translated into actual product performance, into performance parameters such as speed and accuracy that customers require.
Of course, the most important thing is that we need to conduct reliability research and testing! The customer's 24/7 usage scenario is a major concern for businesses. If this doesn't pass, not only will you be on the phone until the middle of the night every day, leading to nervous breakdowns, but everyone from engineers to sales staff will be filled with anxiety! Now that the core components are resolved, let's go all out and solve this problem more comprehensively, and get it right away!
Disclaimer: This article is a reprint. If it involves copyright issues, please contact us promptly for deletion (QQ: 2737591964). We apologize for any inconvenience.
